Mailing-List: contact issues-help@flink.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@flink.apache.org
Received-SPF: pass (athena.apache.org: local policy)
Date: Tue, 5 May 2015 00:19:06 +0000 (UTC)
From: "ASF GitHub Bot (JIRA)" <jira@apache.org>
To: issues@flink.incubator.apache.org
Message-ID: <JIRA.12786158.1427453040000.76674.1430785146651@Atlassian.JIRA>
In-Reply-To: <JIRA.12786158.1427453040000@Atlassian.JIRA>
References: <JIRA.12786158.1427453040000@Atlassian.JIRA>
 <JIRA.12786158.1427453040509@arcas>
Subject: [jira] [Commented] (FLINK-1792) Improve TM Monitoring: CPU
 utilization, hide graphs by default and show summary only
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/FLINK-1792?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14527617#comment-14527617 ] 

ASF GitHub Bot commented on FLINK-1792:
---------------------------------------

Github user bhatsachin commented on a diff in the pull request:

    https://github.com/apache/flink/pull/553#discussion_r29636610
  
    --- Diff: flink-runtime/src/main/scala/org/apache/flink/runtime/taskmanager/TaskManager.scala ---
    @@ -129,6 +130,41 @@ class TaskManager(val connectionInfo: InstanceConnectionInfo,
         override def getValue: Double =
           ManagementFactory.getOperatingSystemMXBean().getSystemLoadAverage()
       })
    +
    +  // Preprocessing steps for registering cpuLoad
    +  // fetch the method to get process CPU load
    +  val getCPULoadMethod: Method = getMethodToFetchCPULoad()
    +
    +  // define the fetchCPULoad method as per the fetched getCPULoadMethod
    +  val fetchCPULoad: (Any) => Double = if (getCPULoadMethod != null) {
    +    (obj: Any) => getCPULoadMethod.invoke(obj).asInstanceOf[Double]
    +  } else {
    +    (obj: Any) => {
    +      log.warning("getProcessCpuLoad method not available in the Operating System Bean" +
    +        "implementation for this Java runtime environment",Thread.currentThread().getStackTrace)
    +      -1
    +    }
    +  }
    +
    +  metricRegistry.register("cpuLoad", new Gauge[Double] {
    +    override def getValue: Double = {
    +      try{
    +        val osMXBean = ManagementFactory.getOperatingSystemMXBean().
    +          asInstanceOf[com.sun.management.OperatingSystemMXBean]
    +        return fetchCPULoad(osMXBean)
    +      } catch {
    +        case t:Throwable => {
    +          if (t.isInstanceOf[java.lang.ClassCastException]){
    +            log.warning("Error casting to OperatingSystemMXBean",t)
    --- End diff --
    
    No, this error is not occurring for each heartbeat. This is in case there is a new error while using the getProcessCpuLoad method. I will remove the ClassCastException and make it Throwable so as to catch all types of errors.


> Improve TM Monitoring: CPU utilization, hide graphs by default and show summary only
> ------------------------------------------------------------------------------------
>
>                 Key: FLINK-1792
>                 URL: https://issues.apache.org/jira/browse/FLINK-1792
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Webfrontend
>    Affects Versions: 0.9
>            Reporter: Robert Metzger
>            Assignee: Sachin Bhat
>
> As per https://github.com/apache/flink/pull/421 from FLINK-1501, there are some enhancements to the current monitoring required
> - Get the CPU utilization in % from each TaskManager process
> - Remove the metrics graph from the overview and only show the current stats as numbers (cpu load, heap utilization) and add a button to enable the detailed graph.


--
This message was sent by Atlassian JIRA
(v6.3.4#6332)